Conflicts of Interest Common in ACC-AHA Practice Guidelines

Download this article's Factoid (PDF & PPT for Gold Subscribers)

Slightly more than half of individuals who participated in the development of clinical practice guidelines for the American College of Cardiology (ACC) and the American Heart Association (AHA) in recent years reported a potential conflict of interest, according to findings published in the March 28, 2011, issue of Archives of Internal Medicine.

But lead author James N. Kirkpatrick, MD, of the University of Pennsylvania (Philadelphia, PA), told TCTMD in a phone interview that the results were actually reassuring. Based on the literature, he said, the researchers had expected the prevalence of conflicts to hit 80%.

In response to the paper, the ACC and AHA promptly issued a joint press release clarifying their policies on conflicts of interest and guideline development. “[T]he data within and the conclusions drawn by the article do not reflect the reality of the guidelines development process today,” they stress, because the study by Dr. Kirkpatrick and colleagues spans 2003 to 2008. A major shift occurred in 2010, they note, when the 2 organizations adopted more stringent rules on industry relationships to align with the Council of Medical Specialty Societies.

Connecting the Dots

In fact, the investigators chose to analyze ACC/AHA guidelines precisely because their high standards made such research possible, Dr. Kirkpatrick said, by outlining disclosures at the end of each document.

For the study, Dr. Kirkpatrick and colleagues evaluated 17 guidelines published jointly by the ACC and AHA between 2003 and 2008. Using disclosure lists, the researchers gathered information on conflicts in 4 categories:

• Receiving a research grant
• Serving on a speaker’s bureau and/or receiving honoraria
• Owning stock
• Being a consultant or member of an advisory board

Participation was quantified in terms of at least 1 instance of participation in 1 guideline by 1 person.

The researchers identified 658 instances of participation by 498 individuals. In all, 56% of those who participated in writing ACC/AHA guidelines had some sort of conflict, most commonly serving as a consultant or advisory board member. Each person took part in an average of 1.31 guidelines (range, 1-7), but only 21% were involved in 2 or more. The proportion of participants with conflicts for each guideline ranged widely, from 13% to 87%.

Guideline committee members were more likely to possess a conflict than were peer reviewers (63% vs. 51%; P = 0.006), as were chairs, co-chairs, and first authors compared with other participants (81% vs. 55%; P = 0.03). The likelihood of having relationships rose in conjunction with increasing participation (P < 0.001 for both the presence and number of conflicts).

In terms of individual guidelines, the number of conflicts ranged from a high of 68 in the STEMI guidelines published in 2006 to a low of 15 in the echo (2003) and intracerebral hemorrhage (2007) guidelines.

A total of 510 commercial companies were mentioned in disclosures. The mean number of companies per guideline was 38 but ranged between 2 and 242. In contrast, only 18 noncommercial entities, such as hospitals, foundations, and universities, were listed. A few companies were most reported in multiple guidelines, with 1 company, Medtronic (Minneapolis, MN) being the most reported in 7 of 17 guidelines.

Findings Raise Alarm Bells, to Varying Degrees

The prevalence of conflicts “is a cause for concern,” the authors note, especially “given the fact that many of the newest ACC/AHA guideline recommendations are based more on expert opinion than on clinical trial data.”

An editorial accompanying the paper by Steven E. Nissen, MD, of the Cleveland Clinic (Cleveland, OH), is more emphatic. “Clinical practice guidelines play an enormously important role in society and the practice of medicine,” he asserts. “Increasingly, government, the public, and the media use [these guidelines] as a benchmark to gauge the quality of medical practice for both hospitals and individual physicians.”

In light of the need for integrity and reliability in practice guidelines, Dr. Nissen says, “[t]he depth and breadth of industry relationships reported in this article are extraordinary.”

Larry S. Dean, MD, of the University of Washington Medicine Regional Heart Center (Seattle, WA), and president of the Society for Cardiovascular Angiography and Interventions (SCAI), told TCTMD in a telephone interview that he was encouraged and pleasantly surprised by the results.

The Institute of Medicine has recently published a report on guideline development that stipulates fewer than half of people contributing to these documents can have conflicts of interest, he reported. “If we’re at 56% through 2008, we could get below that magic number of 50% fairly easily. I’d be interested to see in 2012, because it takes some time to percolate and write these guidelines, whether we’ve met that requirement,” Dr. Dean said. “I think it would be fairly easy to do actually, based on what the ACC and AHA have already done.”

Dr. Kirkpatrick noted that the level of concern depends on who you talk to. He pointed to the finding that 44% of people in the study reported no conflicts, which contradicts the claims that such relationships are a necessary evil in guideline development.

Dr. Dean, on the other hand, expressed some doubt that there exists a sufficient pool of people without conflicts. “I think that’s a little superficial, because not everyone on the writing committee would be able to do different parts of the job,” he said. “They may not have the skill set to do that. You’re just assuming a body is a body, basically, and you have to be careful about that.”

The bottom line, Dr. Kirkpatrick acknowledged, is that “we really don’t know,” how much the conflicts matter, “but I don’t know that we could prove that. Our study certainly wasn’t able to get to that level of detail.

“There’s a classic ethical study suggesting that everyone feels conflicts of interest are bad, but they feel that they’re not affected by their conflicts of interest,” he continued. “You can’t have it both ways.”

Even more concerning is how the findings are perceived, given “the fact that trust in physicians is eroding. When we have a lot of conflicts, then we end up potentially eroding it even more,” Dr. Kirkpatrick said. Despite the possibility that many conscientious physicians are able to set aside bias when producing guidelines, “in the public’s perception this may not matter that much. And without any data showing that people are able to put these aside and produce something that is not biased, we’re sort of left wondering.”

Similarly, Dr. Nissen argues in his editorial that even if the influence of conflicts is unknown, their presence alone “raises appropriate concerns.”

“The revelations reported in the current article highlight troubling concerns that must be urgently addressed. If we fail as a profession to police our [guideline-making] process, the credibility of evidence-based medicine will suffer irreparable harm,” he concludes.

Cure for the Common Conflict?

Disclosures are the most obvious remedy to conflicts but may not fully fix the situation, Dr. Kirkpatrick and colleagues note.

“While publishing the tables of disclosures theoretically gives the readers and reviewers of the guidelines the same opportunity to evaluate the potential for bias in the overall recommendations, scrutinizing several pages of [conflicts of interest] in small print at the end of the guideline or reading the transcript of discussion about a recommendation may not be a high priority for busy clinicians,” they explain. “Furthermore, transcripts of the meetings are not available, and which participant(s) advocated specific recommendations cannot be discerned.”

Dr. Nissen stresses that disclosures are inadequate, because they do nothing to guarantee scientific independence of the resulting guidelines. “The deliberations in writing [guidelines] take place in secret,” he writes. “Accordingly, we will never know the extent to which financial relationships affected the internal discussion and deliberations leading to the final . . . recommendations.”

Other options proposed by the study authors include mandating that committee members have no conflicts related to the guideline they are creating or, if they do, then participants must abstain from voting but may contribute to discussion. Individuals also could divest their financial interests prior to participating.

The ACC and AHA’s refined policies on industry relationships and guideline development for 2010 feature several recent additions:

• Chairs of writing committees may not have any relevant ties to industry
• Apart from the chair, a majority of committee members should also be free of such relationships
• Committee members may not draft recommendations or text, nor vote on any issues that involve any of their industry ties
• Members of the final approving bodies of both organizations must also abstain from voting if they have a relevant relationship with industry

Although the importance of guidelines has been debated, Dr. Kirkpatrick said, “I really don’t think we should throw the baby out with the bath water. And I also really do not want the good work that the ACC has done in dealing with disclosure and being at the forefront of [these issues] to get overshadowed, because it has done a great job among all the subspecialty societies and served as a model,” he commented, adding, “I’m hoping that other societies will follow its lead in all of this.”

Dr. Dean confirmed that, in his time as SCAI president over the past year, the ACC and AHA are following through on their new policies. “When we at SCAI are requested to provide members to the writing committees, it’s always been made very clear that the particular position requires no conflicts of interest,” he said.

As to whether the various remedies for conflicts are necessary or effective, Dr. Dean noted, “It’s a little easy to sit in a room and come up with ideals. I think that’s a great thing to give us direction and improve processes, but when it gets down to implementation of these things, that’s where it gets really complicated.”

Even so, he concluded, the IOM, professional societies, and research such as this, “push us toward a better process, and that’s admirable and good.”

 

---

Sources:
1. Mendelson TB, Meltzer M, Campbell EG, et al. Conflicts of interest in cardiovascular clinical practice guideline. Arch Intern Med. 2011;171:577-585.

2. Nissen SE. Can we trust cardiovascular practice guidelines? Arch Intern Med. 2011;171:584-585.

3. American College of Cardiology/American Heart Association. Institute of Medicine makes recommendations on producing trustworthy guidelines [press release]. March 28, 2011.

 

 

Related Stories:

• Physician-Industry Relationships Decreasing but Still Prevalent
• Patients, Research Participants Prefer Physician Disclosure of Financial Ties
• Financial Conflicts Trouble Potential Research Participants